# Eval

> Run scenarios through targets, score observations, and persist a safe report projection.

- Path: `Modules > Eval - Evaluation Runtime`
- Human: https://looprig.com/docs/modules/eval
- Machine index: https://looprig.com/llms.txt

## Repository

| Field | Value |
| --- | --- |
| Repository | `github.com/looprig/eval` |
| Version | `v0.2.0` |
| GitHub | [looprig/eval](https://github.com/looprig/eval) |

## Description

Run scenarios through targets, score observations, and persist a safe report projection.

## Where it fits

Eval is useful on its own when an application needs to run cases against a target, score observations, and write redacted reports. Within Looprig, it evaluates [Inference](/docs/modules/inference.md) targets and can observe [Harness](/docs/modules/harness.md) behavior. Eval defines the evaluation domain; Pluto packages that domain into a complete qualification product.

## Dependencies

- [Core](/docs/modules/core.md)
- [Inference](/docs/modules/inference.md)

## Dependents

None.